abstention concept

Definition	The verdict a group receives when it has too few rows or too few queries to score. It is reported as a verdict, not dropped. Chapter 10. Also abstain.
Example	A group with 12 rows against a minimum of 50 is reported as abstained rather than scored, and the count of abstentions is printed beside the verdict.
Book	Data Mining as Observation, draft 0.2, commit `f3914f0`; entry id `abstention`, kind concept.
Status	no ledger row names this entry. Corrections: none recorded.
Defining equation	Book equation 10.7.
Assumptions and scope	A stratum with too few rows or too few queries to score receives the verdict abstain, which is reported under a registered cause and excluded from the minimum. Abstain is not a pass, and when no stratum is eligible the verdict itself is abstain. Raising the eligibility bar can only remove strata from the minimum and so can only raise the verdict, which is why every abstention is printed beside the verdict rather than dropped.
Prior art	none recorded
Evidence	`turboquant-pro/docs/HUBNESS_PRIMER.md:140-160`, `turboquant-pro/docs/STRATA_RFC.md:24-98`, `lean/DataMiningAsObservation/Abstention.lean`
Reviewed	not yet reviewed; generated 2026-09-10 from records at the commits on the provenance page.

A verdict per group, with the group too thin to score reported as an abstention rather than dropped.

Equation

Book equation 10.7.

\[\text{verdict}=\min_{i:\ n_i\ge n_{\min},\ |Q_i|\ge q_{\min}}\ \mathrm{score}_i,\qquad \text{ABSTAIN otherwise}.\]

Conditions

A stratum with too few rows or too few queries to score receives the verdict abstain, which is reported under a registered cause and excluded from the minimum. Abstain is not a pass, and when no stratum is eligible the verdict itself is abstain.
Raising the eligibility bar can only remove strata from the minimum and so can only raise the verdict, which is why every abstention is printed beside the verdict rather than dropped.

Conditions are curated in entries.toml rather than read from a record.

Ledger

none

First stated

The strata design in turboquant-pro, turboquant-pro/docs/STRATA_RFC.md:24-98, and chapter 10 section 10.4 of Data Mining as Observation.

Measurements

Where the book states it	Numbers, as the book’s sources table records them	Source
chapter 10 section 10.3	abstain below 2.5 k	`turboquant-pro/docs/HUBNESS_PRIMER.md:140-160`
chapter 10 section 10.4	area map, boundary rule, hash, refuse not warn, intra and transit counts, area classes, abstention rule	`turboquant-pro/docs/STRATA_RFC.md:24-98`

Failures and corrections

none

Invariance envelope

none declared

Machine checked

lean/DataMiningAsObservation/Abstention.lean, theorems abstain_not_passes, verdict_eq_none_iff, passes_verdict_iff, inf_le_of_subset, at observation-data-mining f3914f0; what the check covers is stated in the book’s appendix C.

Used in

Data Mining as Observation chapters 0, 10, 11, 14.

min-over-strata; anti-hub; certificate; coverage.

Status

Generated 2026-09-10 by encyclopedia/generate.py; book at observation-data-mining f3914f0; the commit of every record is listed in the encyclopedia’s provenance.

accuracy →